ttsds/bark_small – Run with an API on Replicate

ttsds / bark_small

The small version of the Bark model by Suno.

Cold

Public
227 runs
A100 (80GB)
GitHub
Weights
License

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

Video Player is loading.

Current Time 00:00:000

Duration 00:00:000

Loaded: 0%

Stream Type LIVE

Remaining Time 00:00:000

language

*string

An enumeration.

text

*string

Shift + Return to add a new line

With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.

speaker_reference

*file

Run this model in Node.js with one line of code:

npx create-replicate --model=ttsds/bark_small

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run ttsds/bark_small using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "ttsds/bark_small:e0e4df86d3dd484bcb7b693077929005d4222d1bbdccf7afed3eccee06b0bbea",
  {
    input: {
      text: "With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.",
      language: "en",
      speaker_reference: "https://replicate.delivery/pbxt/MNFXdPaUPOwYCZjZM4azsymbzE2TCV2WJXfGpeV2DrFWaSq8/example_en.wav"
    }
  }
);

// To access the file URL:
console.log(output.url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run ttsds/bark_small using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "ttsds/bark_small:e0e4df86d3dd484bcb7b693077929005d4222d1bbdccf7afed3eccee06b0bbea",
    input={
        "text": "With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.",
        "language": "en",
        "speaker_reference": "https://replicate.delivery/pbxt/MNFXdPaUPOwYCZjZM4azsymbzE2TCV2WJXfGpeV2DrFWaSq8/example_en.wav"
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run ttsds/bark_small using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "ttsds/bark_small:e0e4df86d3dd484bcb7b693077929005d4222d1bbdccf7afed3eccee06b0bbea",
    "input": {
      "text": "With tenure, Suzie\'d have all the more leisure for yachting, but her publications are no good.",
      "language": "en",
      "speaker_reference": "https://replicate.delivery/pbxt/MNFXdPaUPOwYCZjZM4azsymbzE2TCV2WJXfGpeV2DrFWaSq8/example_en.wav"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

Video Player is loading.

Current Time 00:00:000

Duration 00:00:000

Loaded: 0%

Stream Type LIVE

Remaining Time 00:00:000

{
  "completed_at": "2025-01-27T11:08:52.457579Z",
  "created_at": "2025-01-27T11:06:43.837000Z",
  "data_removed": false,
  "error": null,
  "id": "5gq40x1wznrm80cmmxrt9ahprw",
  "input": {
    "text": "With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.",
    "language": "en",
    "speaker_reference": "https://replicate.delivery/pbxt/MNFXdPaUPOwYCZjZM4azsymbzE2TCV2WJXfGpeV2DrFWaSq8/example_en.wav"
  },
  "logs": "Extracting semantics...\nTokenizing semantics...\nCreating coarse and fine prompts...\n  0%|          | 0/768 [00:00<?, ?it/s]\n  0%|          | 1/768 [00:00<02:24,  5.30it/s]\n  7%|▋         | 51/768 [00:00<00:03, 216.88it/s]\n 13%|█▎        | 101/768 [00:00<00:02, 323.91it/s]\n 20%|█▉        | 152/768 [00:00<00:01, 387.07it/s]\n 26%|██▋       | 202/768 [00:00<00:01, 422.59it/s]\n 33%|███▎      | 252/768 [00:00<00:01, 445.04it/s]\n 39%|███▉      | 303/768 [00:00<00:01, 462.92it/s]\n 46%|████▌     | 354/768 [00:00<00:00, 474.85it/s]\n 53%|█████▎    | 404/768 [00:00<00:00, 482.02it/s]\n 59%|█████▉    | 455/768 [00:01<00:00, 487.87it/s]\n 66%|██████▌   | 506/768 [00:01<00:00, 493.62it/s]\n 73%|███████▎  | 557/768 [00:01<00:00, 496.55it/s]\n100%|██████████| 594/594 [00:01<00:00, 496.55it/s]\n100%|██████████| 594/594 [00:01<00:00, 431.53it/s]\n  0%|          | 0/30 [00:00<?, ?it/s]\n  3%|▎         | 1/30 [00:00<00:03,  7.93it/s]\n  7%|▋         | 2/30 [00:00<00:03,  8.30it/s]\n 10%|█         | 3/30 [00:00<00:03,  8.45it/s]\n 13%|█▎        | 4/30 [00:00<00:03,  8.54it/s]\n 17%|█▋        | 5/30 [00:00<00:02,  8.59it/s]\n 20%|██        | 6/30 [00:00<00:02,  8.59it/s]\n 23%|██▎       | 7/30 [00:00<00:02,  8.64it/s]\n 27%|██▋       | 8/30 [00:00<00:02,  8.67it/s]\n 30%|███       | 9/30 [00:01<00:02,  8.70it/s]\n 33%|███▎      | 10/30 [00:01<00:02,  8.72it/s]\n 37%|███▋      | 11/30 [00:01<00:02,  8.73it/s]\n 40%|████      | 12/30 [00:01<00:02,  8.73it/s]\n 43%|████▎     | 13/30 [00:01<00:01,  8.74it/s]\n 47%|████▋     | 14/30 [00:01<00:01,  8.73it/s]\n 50%|█████     | 15/30 [00:01<00:01,  8.73it/s]\n 53%|█████▎    | 16/30 [00:01<00:01,  8.74it/s]\n 57%|█████▋    | 17/30 [00:01<00:01,  8.75it/s]\n 60%|██████    | 18/30 [00:02<00:01,  8.75it/s]\n 63%|██████▎   | 19/30 [00:02<00:01,  8.75it/s]\n 67%|██████▋   | 20/30 [00:02<00:01,  8.69it/s]\n 70%|███████   | 21/30 [00:02<00:01,  8.70it/s]\n 73%|███████▎  | 22/30 [00:02<00:00,  8.72it/s]\n 77%|███████▋  | 23/30 [00:02<00:00,  8.73it/s]\n 80%|████████  | 24/30 [00:02<00:00,  8.74it/s]\n 83%|████████▎ | 25/30 [00:02<00:00,  8.70it/s]\n 87%|████████▋ | 26/30 [00:02<00:00,  8.72it/s]\n 90%|█████████ | 27/30 [00:03<00:00,  8.71it/s]\n 93%|█████████▎| 28/30 [00:03<00:00,  8.67it/s]\n 97%|█████████▋| 29/30 [00:03<00:00,  8.69it/s]\n100%|██████████| 30/30 [00:03<00:00,  8.76it/s]",
  "metrics": {
    "predict_time": 5.463757617,
    "total_time": 128.620579
  },
  "output": "https://replicate.delivery/xezq/WOBvQ5Q2JlLvJhj1Erj7RUvpq585eQnFCbe31ZYgvMLEtRJUA/test_pred.wav",
  "started_at": "2025-01-27T11:08:46.993821Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bsvm-3iproag4scqrqdwypbylyhpjd7z6shcm6f6p32sktzqokxd2cufq",
    "get": "https://api.replicate.com/v1/predictions/5gq40x1wznrm80cmmxrt9ahprw",
    "cancel": "https://api.replicate.com/v1/predictions/5gq40x1wznrm80cmmxrt9ahprw/cancel"
  },
  "version": "e0e4df86d3dd484bcb7b693077929005d4222d1bbdccf7afed3eccee06b0bbea"
}

Generated in

5.5 seconds

Tweak it ShareReport

Extracting semantics...
Tokenizing semantics...
Creating coarse and fine prompts...
  0%|          | 0/768 [00:00<?, ?it/s]
  0%|          | 1/768 [00:00<02:24,  5.30it/s]
  7%|▋         | 51/768 [00:00<00:03, 216.88it/s]
 13%|█▎        | 101/768 [00:00<00:02, 323.91it/s]
 20%|█▉        | 152/768 [00:00<00:01, 387.07it/s]
 26%|██▋       | 202/768 [00:00<00:01, 422.59it/s]
 33%|███▎      | 252/768 [00:00<00:01, 445.04it/s]
 39%|███▉      | 303/768 [00:00<00:01, 462.92it/s]
 46%|████▌     | 354/768 [00:00<00:00, 474.85it/s]
 53%|█████▎    | 404/768 [00:00<00:00, 482.02it/s]
 59%|█████▉    | 455/768 [00:01<00:00, 487.87it/s]
 66%|██████▌   | 506/768 [00:01<00:00, 493.62it/s]
 73%|███████▎  | 557/768 [00:01<00:00, 496.55it/s]
100%|██████████| 594/594 [00:01<00:00, 496.55it/s]
100%|██████████| 594/594 [00:01<00:00, 431.53it/s]
  0%|          | 0/30 [00:00<?, ?it/s]
  3%|▎         | 1/30 [00:00<00:03,  7.93it/s]
  7%|▋         | 2/30 [00:00<00:03,  8.30it/s]
 10%|█         | 3/30 [00:00<00:03,  8.45it/s]
 13%|█▎        | 4/30 [00:00<00:03,  8.54it/s]
 17%|█▋        | 5/30 [00:00<00:02,  8.59it/s]
 20%|██        | 6/30 [00:00<00:02,  8.59it/s]
 23%|██▎       | 7/30 [00:00<00:02,  8.64it/s]
 27%|██▋       | 8/30 [00:00<00:02,  8.67it/s]
 30%|███       | 9/30 [00:01<00:02,  8.70it/s]
 33%|███▎      | 10/30 [00:01<00:02,  8.72it/s]
 37%|███▋      | 11/30 [00:01<00:02,  8.73it/s]
 40%|████      | 12/30 [00:01<00:02,  8.73it/s]
 43%|████▎     | 13/30 [00:01<00:01,  8.74it/s]
 47%|████▋     | 14/30 [00:01<00:01,  8.73it/s]
 50%|█████     | 15/30 [00:01<00:01,  8.73it/s]
 53%|█████▎    | 16/30 [00:01<00:01,  8.74it/s]
 57%|█████▋    | 17/30 [00:01<00:01,  8.75it/s]
 60%|██████    | 18/30 [00:02<00:01,  8.75it/s]
 63%|██████▎   | 19/30 [00:02<00:01,  8.75it/s]
 67%|██████▋   | 20/30 [00:02<00:01,  8.69it/s]
 70%|███████   | 21/30 [00:02<00:01,  8.70it/s]
 73%|███████▎  | 22/30 [00:02<00:00,  8.72it/s]
 77%|███████▋  | 23/30 [00:02<00:00,  8.73it/s]
 80%|████████  | 24/30 [00:02<00:00,  8.74it/s]
 83%|████████▎ | 25/30 [00:02<00:00,  8.70it/s]
 87%|████████▋ | 26/30 [00:02<00:00,  8.72it/s]
 90%|█████████ | 27/30 [00:03<00:00,  8.71it/s]
 93%|█████████▎| 28/30 [00:03<00:00,  8.67it/s]
 97%|█████████▋| 29/30 [00:03<00:00,  8.69it/s]
100%|██████████| 30/30 [00:03<00:00,  8.76it/s]

Run time and cost

This model costs approximately $0.012 to run on Replicate, or 83 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia A100 (80GB) GPU hardware. Predictions typically complete within 9 seconds.

Readme

Bark by Suno

The small version of the Bark model as released by Suno

Disclaimer

This is an unofficial implementation. Please refer to the Bark repository for the original code and details. By using this model, you agree to the terms stated at the link above, which could change at any time. Make sure to comply to these terms before using the model.